Chapter 3 Kinetics and Dissociation Constants (pKa) of Polyamines of Importance in Post-Combustion Carbon Dioxide (CO2) Capture Studies F. Khalili,† A. V. Rayer,† A. Henni,*,† A. L. L. East,‡ and P. Tontiwachwuthikul† †Industrial Systems Engineering and Applied Science, (International Test Centre for Carbon Capture), University of Regina, Regina, SK S4S 0A2, Canada ‡Department of Chemistry and Biochemistry, University of Regina, Regina, SK S4S 0A2, Canada *E-mail: [email protected] Pseudo-first-order overall rate constants for the loss of CO2 via reaction with different types of amines were measured using a stopped-flow technique at 298.15 K. Polyamines and cyclic amines were found to have higher reaction rates than linear primary and secondary amines. Therefore, six aqueous cyclic polyamine solutions were studied at (298.15 to 313.15) K over a concentration range of (20 to 120) mol·m-3 using. The overall reaction orders were calculated using the empirical Publication Date (Web): May 3, 2012 | doi: 10.1021/bk-2012-1097.ch003 power law kinetics and were found to be fractional in order, Downloaded by UNIV OF REGINA on January 24, 2013 | http://pubs.acs.org for practically all studied cyclic polyamines. The overall rate constants were fitted with the Crooks-Donnellan termolecular rate expression to determine elementary rate constants. In addition, the dissociation constants (pKa) were determined using the potentiometric titration method at (298, 303, 313 and 323) K, and predicted using quantum chemistry techniques (IEFPCM continuum solvation model). A trend was found for the variation of the pKa with the addition of different radical groups to the cyclic base molecules. Computational techniques tested for the prediction of pKa involved B3LYP and MP2 levels of electronic structure theory, the addition of an explicit water molecule inside the continuum cavity, and a special scaling of © 2012 American Chemical Society In Recent Advances in Post-Combustion CO2 Capture Chemistry; Attalla, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012. the cavity radii for the ions. The procedure developed in this study reduced the error found in a previous technique for cyclic amines by 62%. 1. Introduction In order to satisfy environmental combustion standards, acid gases such as CO2 and H2S must be removed from gaseous streams in refining, and in chemical and gas associated production plants. Natural gas and synthetic fuels such as coal gasification and shale oil cover broad ranges of both acid gas composition and pressure (1). The most common solvents used to capture CO2 are the alkanolamines such as the monoethanolamine (MEA), diethanolamine (DEA) and n-methyldiethanol amine (MDEA). These amines are used commercially in post-combustion CO2 capture as aqueous solution or in aqueous organic medium or in combination with aqueous potassium carbonate solutions (2). CO2 capture by chemical absorption using aqueous solution of amine uses absorper abd stripper units. This technology needs to overcome the challenge of reducing the energy, the environmental impact and the capture cost. Finding a better solvent is the path to solve these issues. High cyclic capacity, fast absorption rate, high equilibrium temperature sensitivity and low enthalpy of absorption are some of the factors to be considered in the selection of the best solvent. Both capital and capture costs of CO2 removal depend on the CO2 absorption and desorption rates. Solvents with fast reaction rates can reduce the height of the packing required in both the absorber and stripper. Energy in the stripper can be saved by faster solvents and by achieving a closer equilibrium in the absorber (3). Since 1960, various primary, secondary and tertiary amines were studied for their reaction rate with CO2 to find out faster solvents and many studies were published in the literature, and reviewed by Blauwhoff et al. (4), Versteeg et al. (5), and more recently by Vaidya et al. (6) In this work, a screening study was performed on different types of amines (primary, secondary, tertiary, cyclic and polyamines) using a well established stopped-flow procedure in order to find the best solvents based on their kinetics rates at 298.15 K. Six cyclic Publication Date (Web): May 3, 2012 | doi: 10.1021/bk-2012-1097.ch003 aqueous polyamines were studied at (298.15 to 313.15) K over a concentration Downloaded by UNIV OF REGINA on January 24, 2013 | http://pubs.acs.org range of (20 to 120) mol·m-3. The bascity of the solvent, quantified by the pKa of its conjugate acid, is an important fundamental property which affects the kinetics and possibly the mechanism of the capture process (3–6). A linear relationship between the pKa of an acid or base with its catalytic effect on the reaction rate was reported by Brønsted et al. (4) Many literature studies reported on a Brønsted relationship between the rate constant of the reaction of amines with CO2 and the basicity of such amines (4–8). The following Brønsted relationship was reported by Versteeg et al. (5)for aqueous primary and secondary alkanolamines: and for tertiary amines (5): 44 In Recent Advances in Post-Combustion CO2 Capture Chemistry; Attalla, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012. Because of the essential need for pKa values, it would be useful to predict the aqueous pKa of new amines by a prior means. One direct way is the calculation of such values using a widely popular quantum chemistry program (Gaussian 03) with a continum solvation model, in which the solvent is approximated as a dielectric continuum. A study of this kind was reported by da Silva and Svendsen (9), and improved by our previous work to predict the aqueous pKa values within ± 1 (10). The procedure and the choice of conformers for the studied alkanolamines were recently published. 2. Experimental Setup 2.1. Determination of Chemical Kinetics A stopped flow technique was used for the direct measurement of pseudo first-order kinetics, k0, for different aqueous and non-aqueous diamines, as well as primary, secondary and tertiary alkanolamines. The experimental setup is a standard SF-51 stopped flow unit from Hi-Tech Scientific Ltd., UK. It is an assembly of four major units; a sample-handling unit, a conductivity-detection cell, an A/D converter and a microprocessor. The sample-handling unit is comprised of a stainless steel case which provides support and an enclosure for the sample flow circuit. Schematics of the sample handling unit is shown in Figure 1. The entire flow circuit, with the exception of the stop/waste syringe, is enclosed in a thermostat and maintained at a constant temperature by an external water bath within ± 0.1 K. The front panel of the sample handling unit displays a temperature indicator with a resolution of 0.1 K and an air pressure indicator. A pneumatic air supply is used to control the movement of drive plate located at the bottom of the internal syringes that contains the CO2 solution and amine solution. During an experimental run, a fresh solution of CO2 is loaded into one syringe Publication Date (Web): May 3, 2012 | doi: 10.1021/bk-2012-1097.ch003 and a fresh solution of amine is loaded into the other. Downloaded by UNIV OF REGINA on January 24, 2013 | http://pubs.acs.org CO2 solutions were prepared by bubbling research grade CO2 for at least half an hour through the desired medium with water for an aqueous solution and methanol or ethanol for a non-aqueous solution. The concentration of CO2 in the liquid medium was measured in a gas chromatograph (GC-6890 from Agilent). It was then diluted with the chosen media to keep the CO2 solution at least 10 times lower than the amine solution in order to achieve pseudo first-order conditions. The pseudo first-order rate constants of the aqueous solution of EDA obtained for different concentrations are compared with previously published data (11) in Figure 2. A reproducibility of 4% (absolute average deviation of 15 sets of k0 values from their mean value) and an estimated uncertainty of 5% (absolute average deviation of the obtained value from the literature value) were observed when compared to published data. By fitting the empirical power-law kinetics to the data of the experimentally found pseudo first order constants for CO2 as shown in Figure 2, the reaction order of the amine was determined. 45 In Recent Advances in Post-Combustion CO2 Capture Chemistry; Attalla, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012. Figure 1. Schematic diagram of stopped flow instrument Publication Date (Web): May 3, 2012 | doi: 10.1021/bk-2012-1097.ch003 Downloaded by UNIV OF REGINA on January 24, 2013 | http://pubs.acs.org Figure 2. Pseudo first order reaction for (EDA + H2O) solution. ♦, 298.15 K; ◊, 298.15 K11; ▴, 303.15 K; ▵, 303.15 K11; ▪, 308.15 K; □, 308.15 K11; ●, 313.15 K; ○, 313.15 K11; --- Power law kinetics. 46 In Recent Advances in Post-Combustion CO2 Capture Chemistry; Attalla, M.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012. 2.2. Determination of the Dissociation Constant (pKa) 0.01 M aqueous solutions of amines were prepared using deionized double distilled water. The solution was maintained at the experimental temperature and blanketed with a slow stream of nitrogen. The amine solutions (50 mL) were titrated with 0.1M HCl. Twenty equal portions (0.5 mL) of the titrant were added to the solution and the pH value was read when the equilibrium was reached. The pKa values were determined using the Albert and Serjeant procedure (12). pKa values of piperazine (PZ), from the experimental work are discussed and the same procedure was used to calculate the pKa values of other cyclic amines.